Enhancement of Serum- and Platelet-derived Growth Factor-induced Cell Proliferation by Necl-5/Tage4/Poliovirus Receptor/CD155 through the Ras-Raf-MEK-ERK Signaling*

Necl-5/Tage4/poliovirus receptor/CD155 has been shown to be the poliovirus receptor and to be up-regu-lated in rodent and human carcinoma. We have found previously that mouse Necl-5 regulates cell motility. We show here that mouse Necl-5 is furthermore involved in the regulation of cell proliferation. Studies using a specific antibody against Necl-5 and a dominant negative mutant of Necl-5 revealed that Necl-5 enhanced the se-rum-induced proliferation of NIH3T3, Swiss3T3, and mouse embryonic fibroblast cells. Necl-5 enhanced the serum-induced activation of the Ras-Raf-MEK-ERK signaling, up-regulated cyclins D2 and E, and down-regu-lated p27 Kip1 , eventually shortening the period of the G 0 /G 1 phase of the cell cycle in NIH3T3 cells. Necl-5 simi- larly enhanced the platelet-derived growth factor-in-duced activation of the Ras-Raf-MEK-ERK signaling and shortened the period of the G 0 /G 1 phase of the cell cycle in NIH3T3

Necl-5/Tage4/poliovirus receptor (PVR) 1 /CD155 is an immunoglobulin (Ig)-like molecule having a domain structure consisting of one extracellular region with three Ig-like loops, one transmembrane region, and one cytoplasmic region (1)(2)(3)(4). Human PVR/CD155 was originally identified as the human PVR (1,2), whereas rodent Tage4 was originally identified as the product of a gene overexpressed in rat and mouse colon carcinoma (3,4). PVR/CD155 has also been shown to be overexpressed in human colorectal carcinoma and malignant glioma (5,6). The PVR/CD155 gene has thus far been found only in the primates, and the Tage4 gene has thus far been found only in the rodent, but these genes are likely to be derived from the common ancestor gene (7,8) and tentatively renamed nectinlike molecule-5, Necl-5 (for a review, see Ref. 9). Nectin-like molecules (Necls) have been named for a group of Ig-like molecules whose domain structures are similar to, but slightly different from, those of nectins (9). Nectins are Ca 2ϩ -independent Ig-like cell-cell adhesion molecules that constitute a family of four members, nectin-1, -2, -3, and -4 (for reviews, see Refs. 9 and 10). Nectins form cis-dimers followed by formation of trans-dimers (trans-interaction), eventually causing cell-cell adhesion. Nectins first form cell-cell adhesion where cadherins are recruited, resulting in formation of adherens junctions in epithelial cells and fibroblasts. Nectins are associated with the actin cytoskeleton through afadin, a nectin-and actin filamentbinding protein, as cadherins are associated with the actin cytoskeleton through ␣and ␤-catenins (for a review, see Ref. 11). Necl-5 is one member of the Necl family consisting of five members, Necl-1, -2, -3, -4, and -5 (9). Although they have domain structures similar to those of nectins, they do not directly bind afadin.
Nectin-3 forms not only homo-trans-dimers but also heterotrans-dimers (heterophilic trans-interaction) with either nectin-1 or -2 (9,10). Nectin-4 forms hetero-trans-dimers with nectin-1, but nectin-1 does not form hetero-trans-dimers with nectin-2. These hetero-trans-dimers show much higher cell-cell adhesion activity than the homo-trans-dimers. In contrast to nectins, Necl-5 does not show homophilic cell-cell adhesion activity (12,13). Thus, the role of Necl-5 as the PVR has been established, but its physiological role remained unknown for a long time. Our group and Wimmer's groups (13,14) have independently shown recently that Necl-5 has heterophilic cell-cell adhesion activity selectively with nectin-3. We have moreover found that Necl-5 is up-regulated in NIH3T3 cells transformed by an oncogenic Ras (V12Ras-NIH3T3 cells) (13) and that Necl-5 is functionally associated with integrin ␣ V ␤ 3 at leading edges of migrating cells (15) and enhances serum-and plateletderivedgrowthfactor(PDGF)-inducedcellmotilityinanintegrindependent manner, when Necl-5 does not trans-interact with nectin-3 (15). Studies on the mechanisms of the cell migrating activity of Necl-5 have revealed that the extracellular region of Necl-5 is necessary for directional cell migration, but not for random cell motility. The cytoplasmic region is necessary for both random cell motility and directional cell migration. Necl-5 enhances the serum-induced activation of Cdc42 and Rac, causing formation of filopodia and lamellipodia, respectively, which are necessary for cell motility. The cytoplasmic region of hu-man Necl-5 binds Tctex-1, a subunit of the dynein motor complex (16), which may also be involved in the regulation of cell motility in cooperation with microtubules. Thus, Necl-5 plays key roles not only in poliovirus infection but also in cell motility and is likely to be responsible for enhanced motility of transformed cells. Transformed cells generally show not only enhanced cell motility but also enhanced cell proliferation. We have therefore examined here whether Necl-5 is furthermore involved in regulation of cell proliferation.
Quantification of Necl-5 by Flow Cytometry-For flow cytometry of endogenous Necl-5, the cells were detached with 5 mM EDTA and phosphate-buffered saline, suspended in FACS buffer (1% BSA, 0.02% sodium azide in phosphate-buffered saline), and incubated with the anti-Necl-5-mAb-i or control rat IgG (Jackson Immunoresearch) in FACS buffer. After being washed, the cells were incubated with a fluorescein isothiocyanate-conjugated donkey anti-rat IgG pAb (Chemicon). The cells were then analyzed with a FACSort and Cell Quest software (BD Biosciences).
Assays for Cell Growth and Cell Cycle-The cells examined were seeded at a density of 3.0 ϫ 10 4 and 9.0 ϫ 10 5 cells/well in a 24-well plate and a 10-cm dish, respectively. The plating efficiencies of NIH3T3, Swiss3T3, and MEF cells were about 80, 40, and 50%, respectively.
After 4 h, the cells were washed with DMEM and starved of serum with DMEM containing 0.5% fatty acid-free BSA for 24 or 48 h. After the serum starvation, the quiescent cells were cultured in DMEM containing 10% serum in the presence or absence of 100 g/ml anti-Necl-5 mAb-s or DMEM containing 0.5% fatty acid-free BSA, 20 ng/ml PDGF or 10 ng/ml FGF, and 10 g/ml insulin. For analysis of cell growth curve, the cells were trypsinized and counted using a hemacytometer after various periods of time. For cell cycle analysis by flow cytometry, the cells were detached with 0.05% trypsin and EDTA and fixed in 100% ethanol at 4°C overnight. The cells were stained with 50 g/ml propidium iodide (Wako), 0.1% Nonidet P-40, 0.1% trisodium citrate, and 0.1 mg/ml RNase and analyzed with a FACSort and Cell Quest software (BD Biosciences).
Assays for Up-regulation of Cyclins D2 and E, Down-regulation of p27 Kip1 , and Activation of MEK and ERK-The cells were seeded and cultured as described above. After the serum starvation, the quiescent cells were cultured in DMEM containing 10% serum in the presence or absence of 100 g/ml anti-Necl-5 mAb-s or DMEM containing 0.5% fatty acid-free BSA and 3 ng/ml PDGF. After being washed with ice-cold phosphate-buffered saline, the cells were harvested with prewarmed Laemmli buffer (21) and boiled for 5 min. Protein concentrations of the samples were determined by an RC DC protein assay kit (Bio-Rad) with BSA as a reference protein. The samples were subjected to SDS-PAGE (21), followed by Western blotting using the various Abs and ECL plus Western blotting detection reagents (Amersham Biosciences).

Enhancement of Cell Proliferation by Necl-5-We have
shown previously that Necl-5-⌬CP (Necl-5 of which the cytoplasmic region is deleted) serves as a dominant negative mutant of Necl-5 and inhibits the cell motility enhancing activity of Necl-5 in wild-type and V12Ras-transformed NIH3T3 cells (15). We first obtained NIH3T3 cells stably expressing FLAGtagged Necl-5-⌬CP (Necl-5-⌬CP-NIH3T3 cells) and Zeocin-resistant NIH3T3 cells (Zeo-NIH3T3 cells) as an empty vector control. FLAG-tagged Necl-5-⌬CP was indeed expressed in Necl-5-⌬CP-NIH3T3 cells as estimated by Western blotting (Fig. 1A). Necl-5-⌬CP-NIH3T3, Zeo-NIH3T3, and wild-type NIH3T3 cells were starved of serum for 24 h and then cultured in the presence of serum. The cells continued to proliferate until they became confluent. The growth rate of Zeo-NIH3T3 cells was similar to that of wild-type NIH3T3 cells ( Fig. 2A). However, the growth rate of Necl-5-⌬CP-NIH3T3 cells was much slower than that of Zeo-NIH3T3 and wild-type NIH3T3 cells. Furthermore, the density of Necl-5-⌬CP-NIH3T3 cells at confluence was far less than that of Zeo-NIH3T3 and wild-type NIH3T3 cells. We have shown previously that one of the anti- Necl-5 mAbs, the anti-Necl-5 mAb-s, stimulates the cell motility enhancing activity of Necl-5 (15). We then examined the effect of this mAb on the proliferation of wild-type NIH3T3 cells. The growth rate was faster in the presence of the mAb than in the absence of the mAb (Fig. 2A). The cell density at confluence was higher in the presence of the mAb than in the absence of the mAb. Furthermore, we obtained NIH3T3 cells inducibly expressing FLAG-tagged Necl-5-⌬CP by use of the Tet-Off system (Tet-Off-Necl-5-⌬CP-NIH3T3 cells) (Fig. 1B). The growth rate of Tet-Off-NIH3T3 cells, which were the parental cells of Tet-Off-Necl-5-⌬CP-NIH3T3 cells, was similar to that of wild-type NIH3T3 cells in the presence or absence of doxycycline (Fig. 2, A and B). When FLAG-tagged Necl-5-⌬CP was inducibly expressed in the absence of doxycycline, the growth rate of Tet-Off-Necl-5-⌬CP-NIH3T3 cells was much slower than that of Tet-Off-NIH3T3 and wild-type NIH3T3 cells. When the expression of FLAG-tagged Necl-5-⌬CP was suppressed by doxycycline, the growth rate of Tet-Off-Necl-5-⌬CP-NIH3T3 cells became faster than that in the absence of doxycycline (Fig. 2B). However, this growth rate was slower than that of Tet-Off-NIH3T3 and wild-type NIH3T3 cells because of the expression of FLAG-tagged Necl-5-⌬CP to a small extent which was not completely suppressed by doxycycline (Fig. 1B). Taken together, these results indicate that Necl-5 is involved in the regulation of proliferation of NIH3T3 cells.
We further confirmed by use of other cell lines that Necl-5 is involved in the regulation of cell proliferation. Necl-5 was undetected by Western blotting in Swiss3T3 and MEF cells (data not shown), but it was expressed in these cell lines as estimated by flow cytometry (Fig. 1Ca). Therefore, we obtained Swiss3T3 and MEF cells stably expressing FLAG-tagged Necl-5-⌬CP (Necl-5-⌬CP-Swiss3T3 and Necl-5-⌬CP-MEF cells, respectively). FLAG-tagged Necl-5-⌬CP was indeed expressed in these cell lines as estimated by Western blotting (Fig. 1A). FLAG-tagged Necl-5-⌬CP and the anti-Necl-5 mAb-s showed the essentially similar inhibitory and stimulatory effects, respectively, on the proliferation of Swiss3T3 and MEF cells, although the extents of the inhibitory and stimulatory effects were slightly different among the three cell lines used (Fig. 2,  C and D). These results indicate that Necl-5 is involved in the regulation of cell proliferation.
Shortening the period of the G 0 /G 1 Phase of the Cell Cycle by Necl-5-We next investigated using NIH3T3 cells how Necl-5 enhanced cell proliferation. We first performed flow cytometry to clarify which phase of the cell cycle is enhanced by Necl-5. Necl-5-⌬CP-NIH3T3 and wild-type NIH3T3 cells were starved of serum for 24 h and then cultured in the presence of serum for 18 h. During the 18 h, the number of cells in the G 0 /G 1 phase was decreased less markedly in Necl-5-⌬CP-NIH3T3 cells than in wild-type NIH3T3 cells, and conversely the number of cells in the S phase was increased less markedly in Necl-5-⌬CP-NIH3T3 cells than in wild-type NIH3T3 cells (Fig. 3A, a and b). It took 15 h for the progression of the G 0 /G 1 phase to the S phase of wild-type NIH3T3 cells, whereas it took 21 h for that of Necl-5-⌬CP-NIH3T3 (data not shown). The number of wildtype NIH3T3 cells in the G 0 /G 1 phase was decreased more markedly in the presence of the anti-Necl-5 mAb-s than in the absence of the mAb, and conversely the number of wild-type NIH3T3 cells in the S phase was increased more markedly in the presence of the mAb than in the absence of the mAb (Fig.  3Aa). These results, essentially similar to those obtained for wild-type NIH3T3 cells, were obtained for Zeo-NIH3T3 cells (data not shown). These results indicate that Necl-5 shortens the time of the G 0 /G 1 phase of the cell cycle.
Up-regulation and Down-regulation of Cell Cycle Regulators by Necl-5-Many components have been identified to be in-volved in transition from the G 0 phase to the G 1 phase and progression of the G 1 phase to the S phase (for reviews, see Refs. 23 and 24). Cyclin D is first up-regulated followed by down-regulation of the cyclin-dependent kinase inhibitor p27 Kip1 and up-regulation of cyclin E in this order at the G 1 phase in NIH3T3 cells (25). We then examined the expression levels of these cell cycle regulators in Necl-5-⌬CP-NIH3T3 and wild-type NIH3T3 cells by Western blotting. Cyclins D2 and E were down-regulated more markedly in Necl-5-⌬CP-NIH3T3 cells than in wild-type NIH3T3 cells, whereas they were upregulated more markedly in the presence of the anti-Necl-5-mAb-s than in the absence of the mAb in wild-type NIH3T3 cells (Fig. 4A). In contrast, p27 Kip1 was less down-regulated in Necl-5-⌬CP-NIH3T3 cells than in wild-type NIH3T3 cells, whereas it was more down-regulated in the presence of the anti-Necl-5-mAb-s than in the absence of the mAb in wild-type NIH3T3 cells. These results, essentially similar to those obtained for wild-type NIH3T3 cells, were obtained for Zeo-NIH3T3 cells (data not shown). The results indicate that Necl-5 shortens the time of the G 0 /G 1 phase through regulating the key cell cycle regulators.
Enhancement by Necl-5 of the Serum-induced Activation of the Ras-Raf-MEK-ERK Signaling-It has been shown that the Ras-Raf-MEK-ERK signaling regulates cell cycle regulators and that serum and growth factors induce activation of this signaling in a variety of cell lines (for reviews, see Refs. 26 and 27). We therefore examined whether Necl-5 enhanced the serum-induced activation of each component of this signaling. Consistent with earlier observations (28), serum induced activation of Ras, MEK, and ERK in wild-type NIH3T3 cells (Fig.  4B). The serum-induced activation of each component was weaker in Necl-5-⌬CP-NIH3T3 cells than in wild-type NIH3T3 cells, and the activation was less sustained in Necl-5-⌬CP-NIH3T3 cells compared with wild-type NIH3T3 cells. The serum-induced activation of each component was further enhanced and more sustained in the presence of the anti-Necl-5 mAb-s than in the absence of the mAb. These results, essentially similar to those obtained for wild-type NIH3T3 cells, were obtained for Zeo-NIH3T3 cells (data not shown). Taken together, the results indicate that Necl-5 enhances the serum-  (29,30). We therefore examined whether the PDGF-induced proliferation was regulated by Necl-5. We performed flow cytometry of NIH3T3 cells cultured in the presence of PDGF for 21 h because the time required for the PDGF-induced progression of the G 0 /G 1 phase was much longer than that required for the serum-induced progression. During the 21 h, the number of cells in the G 0 /G 1 phase was decreased less markedly in Necl-5-⌬CP-NIH3T3 cells than in wild-type NIH3T3 cells, and conversely the number of cells in the S phase was increased less markedly in Necl-5-⌬CP-NIH3T3 cells than in wild-type NIH3T3 cells (Fig.  3, Aa (0 h), Ab (0 h), Ba, and Bb). The number of wild-type NIH3T3 cells in the G 2 /M phase was increased more markedly in the presence of the anti-Necl-5-mAb-s than in the absence of the mAb, and conversely the number of wild-type NIH3T3 cells in the S phase was decreased more markedly in the presence of the mAb than in the absence of the mAb (Fig. 3, Aa (0 h) and Ba). The number of wild-type NIH3T3 cells in the G 0 /G 1 phase in the presence of the mAb was not decreased markedly compared with that in the absence of the mAb. This may be because of the cells that had progressed from the G 2 /M phase to the G 1 phase of the second cycle in the presence of the mAb. Essentially similar results were obtained when FGF was used instead of PDGF (data not shown). Results essentially similar to those obtained for wild-type NIH3T3 cells were obtained for Zeo-NIH3T3 cells (data not shown). These results indicate that Necl-5 enhances the PDGF-and FGF-induced cell proliferation by shortening the time of the G 0 /G 1 phase of the cell cycle.
We confirmed that Necl-5 enhanced the PDGF-induced activation of the Ras-Raf-MEK-ERK signaling as described for serum. The PDGF-induced activation of each component was weaker in Necl-5-⌬CP-NIH3T3 cells than in wild-type NIH3T3 cells, and the activation was less sustained in Necl-5-⌬CP-NIH3T3 cells than in wild-type NIH3T3 cells (Fig. 4C). The PDGF-induced activation of each component was rapidly increased and more sustained in the presence of the anti-Necl-5 mAb-s than in the absence of the mAb. In contrast, the phosphorylation at Tyr-857 of PDGF receptor ␤, which is a PDGFinduced major autophosphorylation site (31), was neither reduced in Necl-5-⌬CP-NIH3T3 cells nor elevated by the anti-Necl-5 mAb-s (Fig. 4D). The results, essentially similar to those obtained for wild-type NIH3T3 cells, were obtained for Zeo-NIH3T3 cells (data not shown). Taken together, these results indicate that Necl-5 acts downstream of the PDGF receptor and upstream of Ras and thereby enhances the PDGF-induced activation of the Ras-Raf-MEK-ERK signaling.
Involvement of Up-regulated Necl-5 in Enhanced Proliferation of Transformed Cells-Finally, we examined whether upregulated Necl-5 is involved in the enhanced proliferation of NIH3T3 cells transformed by V12Ras or v-Src (V12Ras-NIH3T3 or v-Src-NIH3T3 cells, respectively). Necl-5 was upregulated not only in V12Ras-NIH3T3 cells (13) but also in v-Src-NIH3T3 cells (Fig. 1Cb). We first obtained V12Ras-NIH3T3 cells stably expressing FLAG-tagged Necl-5-⌬CP (Necl-5-⌬CP-V12Ras-NIH3T3 cells) and Zeocin-resistant V12Ras-NIH3T3 cells (Zeo-V12Ras-NIH3T3 cells) as an empty vector control. We also obtained v-Src-NIH3T3 cells stably expressing FLAG-tagged Necl-5-⌬CP (Necl-5-⌬CP-v-Src-NIH3T3 cells) and Zeocin-resistant v-Src-NIH3T3 cells (Zeo-v-Src-NIH3T3 cells) as an empty vector control. The expression of FLAG-tagged Necl-5-⌬CP in Necl-5-⌬CP-V12Ras-NIH3T3 and Necl-5-⌬CP-v-Src-NIH3T3 cells was confirmed by Western blotting (Fig. 1A). All of these cell lines, V12Ras-NIH3T3, Zeo-V12Ras-NIH3T3, Necl-5-⌬CP-V12Ras-NIH3T3, v-Src-NIH3T3, Zeo-v-Src-NIH3T3, and Necl-5-⌬CP-v-Src-NIH3T3, were starved of serum for 48 h and then cultured in the presence of serum for 15 h. The growth rates of these three cell lines were then compared by flow cytometry. The progression of the G 0 /G 1 phase to the S phase of Zeo-V12Ras-NIH3T3 cells was similar to that of V12Ras-NIH3T3 cells (data not shown). The progression of the G 0 /G 1 phase to the S phase of Zeo-v-Src- NIH3T3 cells was similar to that of v-Src-NIH3T3 cells (data not shown). However, during the 15 h, the number of cells in the G 0 /G 1 phase was decreased less markedly in Necl-5-⌬CP-V12Ras-NIH3T3 cells than in V12Ras-NIH3T3 and Zeo-V12Ras-NIH3T3 cells, and conversely the number of cells in the S phase was increased less markedly in Necl-5-⌬CP-V12Ras-NIH3T3 cells than in V12Ras-NIH3T3 and Zeo-V12Ras-NIH3T3 cells (Fig. 5A, a and b, and data not shown). Essentially similar results were obtained for Necl-5-⌬CP-v-Src-NIH3T3 and Zeo-v-Src-NIH3T3 cells (Fig. 5B and data not  shown). These results indicate that up-regulated Necl-5 is involved at least partly in the enhanced proliferation of transformed cells. DISCUSSION Our preceding papers have shown that Necl-5 regulates cell motility independently, or in cooperation with, nectin-3 (13,15). In the present study, we have first shown by use of wildtype NIH3T3 cells and NIH3T3 cells stably expressing a dominant negative mutant of Necl-5 (Necl-5-⌬CP-NIH3T3 cells) that Necl-5 furthermore regulates the serum-induced proliferation of NIH3T3 cells. We obtained four clones for this cell line, and essentially similar results were obtained for all the cell clones (data not shown). Essentially similar results were obtained for NIH3T3 cell line inducibly expressing Necl-5-⌬CP by use of the Tet-Off system. We have confirmed by use of the specific mAb against Necl-5 that Necl-5 is involved in regulation of the serum-induced proliferation of NIH3T3 cells. We have moreover obtained essentially similar results using Swiss3T3 and MEF cells. All of these results indicate that Necl-5 is involved in the regulation of cell proliferation.
Studies on the mode of action of Necl-5 on its cell proliferation-enhancing activity in NIH3T3 cells have revealed that Necl-5 enhances the serum-induced activation of the Ras-Raf-MEK-ERK signaling, up-regulates cyclins D2 and E, and downregulates p27 Kip1 in NIH3T3 cells. It is difficult to distinguish exactly whether Necl-5 shortens the period of the G 0 phase alone, the G 1 phase alone, or both, but it is likely that Necl-5 shortens the period of both G 0 and G 1 phases. It remains to be elucidated how Necl-5 regulates the cell proliferation in cooperation with serum, but Necl-5 is likely to enhance cell proliferation in cooperation with the growth factor receptor(s), such as the PDGF and FGF receptors, because Necl-5 enhances the PDGF-and FGF-induced proliferation of NIH3T3 cells. Furthermore, Necl-5 enhances the PDGF-induced activation of the Ras-Raf-MEK-ERK signaling but does not affect the PDGFinduced phosphorylation of PDGF receptor ␤ at Tyr-857, which is a major phosphorylation site (31). Therefore, Necl-5 is likely to act downstream of the PDGF receptor and upstream of Ras and thereby enhances the PDGF-induced activation of the Ras-Raf-MEK-ERK signaling, eventually shortening the period of the G 0 /G 1 phase of the cell cycle in NIH3T3 cells.
We have shown previously that Necl-5 is up-regulated in V12Ras-NIH3T3 cells compared with that in wild-type NIH3T3 cells (13). We have shown here that up-regulated Necl-5 by this oncogene is responsible at least partly for the enhanced proliferation of V12Ras-NIH3T3 cells. These results are consistent with our previous observation that up-regulated Necl-5 is responsible at least partly for the enhanced motility of V12Ras-NIH3T3 cells (13,15). We have moreover shown here that up-regulated Necl-5 is responsible at least partly for the enhanced proliferation of another transformed cell line, v-Src-NIH3T3. Taken together, one mechanism of the enhanced movement and proliferation of transformed cells may be upregulated Necl-5.
When nontransformed proliferating cells contact other cells and become confluent, they stop movement and proliferation (for a review, see Ref. 32; see also Ref. 33). This phenomenon was known for a long time as contact inhibition of cell movement and proliferation. Transformation of cells causes loss of this contact inhibition (for reviews, see Refs. 34 and 35). Necl-5 does not show homophilic cell-cell adhesion activity but shows heterophilic cell-cell adhesion activity selectively with nectin-3 (12,13). The role of this heterophilic interaction of Necl-5 with nectin-3 in regulation of cell proliferation remains unknown, but it could be related to the contact inhibition of cell movement and proliferation because as cell density increases, frequency of this heterophilic interaction increases. Further studies are necessary for establishing the physiological and pathological functions of Necl-5 in cell movement and proliferation.